‘RESTING-PLACES OF SAINTS’

The Viking-Age text that sheds light on the location of Cuthbert’s body in the Viking Age is the early-eleventh-century ‘Resting-Places of Saints’, familiar to many under the title Secgan be þam Godes sanctum þe on Engla lande ærost reston or, simply, ‘the Secgan’.Footnote ⁶ The text tells us that Cuthbert lay at Ubbanford, later known as Norham, on the river Tweed: ‘Ðonne resteð sanctus Cuthbertus on þare stowe seo is genemned Ubbanford neh þære éá, þe is genemned Twiode.’Footnote ⁷ Norham, the site of a famous later medieval castle, lies on the northern edge of (what is today) the Northumberland council region, a stone’s throw from (what is now) Scotland. The list took its final form sometime after 1013, when St Florence was interred in Peterborough, but before 1031, by which time the list had to have been entered into Stowe 944.Footnote ⁸ A later update to the list (that is, to the Stowe version), preserved in Cambridge, Corpus Christi College 201, altered ‘seo is genemned Ubbanford neh þære éá, þe is genemned Twiode’, to read ‘þe men hátað Donhólm’.Footnote ⁹ Later in the eleventh century the text was translated into Latin; highlighting the limits of ‘living memory’, the translator appears to have been unfamiliar with the place-name Ubbanford, and incorrectly added ‘vel Dunholm’ to ‘in loco vocatur Ubbanford’, its translation of ‘þare stowe seo is genemned Ubbanford’.Footnote ¹⁰

The date of the ‘Resting-Places of Saints’ presents difficulties for our acceptance of the ‘Flight of Eardwulf’ to Chester-le-Street and suggests that Cuthbert’s move to Durham was later than 995, the date provided for ‘Ealdhun’s Translation’ in the chronological scheme devised by Symeon of Durham. David Rollason, the chief commentator in recent decades on both the ‘Resting-Places of Saints’ and on Symeon of Durham, tried to account for the entry by suggesting that there had been a ninth-century version of the burial list that the extant eleventh-century versions simply copied without update. According to Rollason’s interpretation, the creator of the earlier text had been uninterested in southern saints and had listed only the Northumbrian and ‘midland’ saints we find in the earlier portion of the surviving text; in the early eleventh century (or perhaps before), this ‘first half’ came to be incorporated into the extant version, retaining Cuthbert with his defunct ninth-century resting place unaltered. Rollason pointed out that entries falling in this ‘first half’ are much more likely to be identified with river names (for example, Ubbanford’s localization on the river Tweed) than entries in the ‘second half’ suggesting ‘a different and more elaborate convention’ (p. 62) and that, perhaps, a separate textual tradition lay behind the ‘first half’.

Rollason’s theory is plausible but also, it must be recognized, speculative and insecure. Even if we accepted that there was a ‘first half’ and even if we accepted that it had a separate author (there are surely plenty of other explanations for Rollason’s observation),Footnote ¹¹ there is no independent evidence that it was composed much earlier. A more significant problem is that the creator of the extant text does not, otherwise, verifiably present ‘outdated’ information. Even if Rollason’s ‘first half’ hypothesis is correct and the surviving list is indeed an expanded version of an earlier text (even one that had not excluded southern resting places), that would not be enough to believe fossilized ninth-century entries made their way into the eleventh-century text that we actually have. Even in the ‘first half’, Oswald is located at Gloucester and not Bardney, a move dating to either 909 (Mercian Register) or 906 (ASC MS D).Footnote ¹² Similar anomalous, ‘updates’ would also have occurred for saints Edmund of Bury and Eadburg of Southwell; indeed, of all the English saints, only Cuthbert seems to be ‘misplaced’, a reasonable indication that the problem lies with the eleventh- and twelfth-century Cuthbertine writing at Durham and not the ‘Resting-Places of Saints’. Rollason’s interpretation creates the need for numerous ‘interpolations’ that are not otherwise necessary, including in the allegedly conservative ‘first half’.Footnote ¹³ The need to reconcile the ‘Resting-Places of Saints’ with the eleventh- and twelfth-century traditions would be the only solid reason to believe that the ‘earlier’ ‘first half’ had to be ninth century in date, logic that is open to criticism for circularity. In short, the complications of this Early Core–Interpolation model create more problems than they solve, and the hypothesis is completely unnecessary unless HSC and Symeonic traditions have already convinced us that the ‘Resting-Places of Saints’ must be outdated on the matter of Cuthbert and Norham specifically. On the other hand, the ‘Resting-Places of Saints’ predates our material from Anglo-Norman Durham and verifiably dates to the early eleventh century. Is the latter really reliable enough to necessitate what some might regard as special pleading for one entry in the ‘Resting-Places of Saints’? That is a question that will be examined further below, but first we must recognize that even among Anglo-Norman-era texts there were alternatives to Symeon’s vision of the Cuthbertine Viking Age. Indeed, explicit support for the ‘Resting-Places of Saints’ can even be found elsewhere.

WILLIAM OF MALMESBURY

The ‘Resting-Places of Saints’ is not as isolated as it might first appear. William of Malmesbury’s (†c. 1142) Gesta pontificum Anglorum was completed by 1125 but compiled in preceding years from pre-existing episcopal lists and miscellaneous other sources, including (apparently) pre-Symeonic or proto-Symeonic material from Durham.Footnote ¹⁴ William’s principal value here is that he had access to sources similar to those used by Symeon and, potentially, some of his predecessors at Durham, but reached independent judgements about how to use the material. William applied enormous effort to document the pre-Viking and Viking-Age episcopate of England, and in the case of the Cuthbertine see he managed to find an account of how the diocese of Lindisfarne became the diocese of Durham. According to William, the bodies of the Lindisfarne saints had been moved to the mainland because of the ravaging of the Danes in the ninth century; there was an attempt to move St Cuthbert to Ireland (also attested in the ‘Flight of Eardwulf’), but instead the body was taken to ‘Ubbanford … iuxta amnem Twda’, where it lay until the time of King Æthelred II (reigned 978–1016) – noting however that a ‘correction’ to Cnut (r. 1016–35) appears in later manuscript variations.Footnote ¹⁵ William’s account is very like the ‘Flight of Eardwulf’, but it lacks knowledge of either Eardwulf or Chester-le-Street (though Eardwulf appears, subsequently, in William’s list of bishops).Footnote ¹⁶ William goes on to say that, ‘in the interval’, between the body’s removal and the reign of Æthelred/Cnut, Cuthbert had performed miracles all over England, specifically recounting one concerning Ælfred the Great (iii.129). After the death of Bishop Ealdhun, a certain Bishop Eadmund decides, with the consent of the king, to relocate to Durham. It is worth quoting the passage so that the full detail is appreciated:

The specificity of the detail has the potential to be particularly important, as is the reference to potential sources, ‘as the locals tell’ (‘ut commemorant indigenae loci’); uncertainty in the manuscript tradition about the name of the English king might be interpreted as evidence that William performed the synchronization himself, that the account of Eadmund’s election and episcopate had been supplied without reference to any English king’s reign.Footnote ¹⁸

One could dismiss William’s account, but the broader picture makes that difficult. William of Malmesbury as a potential purveyor of earlier material carries at least equal weight to Symeon, but unlike the latter’s his account corresponds to the testimony of a witness that is verifiably contemporary. That the ‘Resting-Places of Saints’ is Viking-Age in date is not itself, it goes without saying, enough to guarantee its accuracy. Although it is more likely that William’s northern source material is pre- or proto-Symeonic, that is not certain and it is not completely out of the question that William used the ‘Resting-Places of Saints’ and then altered Durham material to accommodate a resting-place at Norham, a complex interpretation that would also require William to reach his own judgement about the date of the burial list as a historical document and use that judgement to alter material coming from the north. The question for the modern historian is how worthwhile is it to create theories as complex as these? Does the evidence we have really demand we go so far? As discussed below, it is much simpler to deny authority to the specifics of HSC’s ‘Flight of Eardwulf’ and Symeon’s ‘Ealdhun’s Translation’ than to explain why the ‘Resting-Places of Saints’ and William are both wrong and why they both get the story wrong in the same specific way.

Moreover, there is even more evidence that Norham, not Chester-le-Street, was the seat of the tenth-century Cuthbertine cult. Before discussing this evidence, however, we need to better understand the potential of the ‘Flight of Eardwulf’ and of ‘Ealdhun’s Translation’, and of HSC and the Symeonic material more generally, as purveyors of accurate information about the Viking Age. It is this that deprives the historian of the right to reject these episodes out of hand.

‘FLIGHT OF EARDWULF’ AND ‘EALDHUN’S TRANSLATION’

HSC, the earliest text to mention the move to Chester-le-Street, is a compilation that was completed sometime in the eleventh or early twelfth century (see below). The relevant episode, the ‘Flight of Eardwulf’, occupies ‘chapter 20’ of the modern edition by South. In the story, Bishop Eardwulf and Abbot Eadred lead the community of Cuthbert during seven years of rootless exile. The wanderers attempt to take passage to Ireland from the mouth of the Derwent (Cumbria) but, lacking heaven’s approval, are compelled to return east, residing at Crayke (Yorkshire) with Abbot Geve, then finally settling at Chester-le-Street (County Durham):

The ‘Flight of Eardwulf’ episode could potentially be a standalone account predating the completion of HSC or it could be an ‘interpolation’ that is later than a ‘core’; what is clear is that the architect of HSC took care to synchronize it with the reigns of kings Guthred (fl. late ninth century) and Alfred the Great (r. 871–99). The ‘Flight’ concludes a narrative sequence also containing ‘Alfred’s Dream’ and the ‘Donation of Guthred’; in the former, Cuthbert’s spirit guides King Alfred to victory over the Danes (chapters 14–18); in the latter, Cuthbert’s power is used to free a certain Guthred from slavery and raise him to the kingship of the ‘army of Danes over the Tyne’ (‘super Tinam ad exercitum Danorum’), whereupon the new ruler rewards Cuthbert’s followers with certain powers of sanctuary and with tenure of the land between the Tees and the Tyne (chapter 13).Footnote ²⁰

HSC does not, it should be repeated, describe the translation from Chester-le-Street to Durham. The first extended account of this second migration, ‘Ealdhun’s Translation’, occurs in Libellus de Exordio [hereafter, LDE], an early-twelfth-century work that builds upon HSC; however, the Ealdhun episode had been referred to briefly in the Annales Lindisfarnenses et Dunelmenses [hereafter ALD], seemingly the preliminary scheme that established much of the dating framework used by LDE (discussed below). LDE is larger and more sophisticated than the comparatively crude HSC, and possesses a more explicit commitment to chronological structure:

The account goes on to relate that the precise resting place favoured by St Cuthbert was only revealed in a vision by a certain Eadmer after several days of fasting and prayer.Footnote ²²

DATING AND ‘EXCAVATING’ HISTORIA DE SANCTO CUTHBERTO

Again, it must be conceded that some of the material in HSC might date to the tenth century. John Hodgson-Hinde in the nineteenth century and Edmund Craster in the 1950s suggested that there had been an early core to HSC, written in the mid-tenth century.Footnote ²³ Their case drew weight from the fact that one of the three extant manuscripts of the text (Cambridge, University Library, Ff. 27) terminates with the tenth-century King Edmund’s visit to St Cuthbert’s church; the other manuscripts (including the older version in Oxford, Bodleian, Bodley 596) incorporate material from the reign of Æthelred (r. 978–1016) and conclude in the reign of Cnut (r. 1016–35), leaving a potentially significant chronological lacuna in the second half of the tenth century. Perhaps shorn of a few ‘interpolations’, Craster suggested, HSC could be understood as ‘originally’ ending with Edmund’s visit; subsequently, another scribe added material relating to the reigns of Æthelred and Cnut, interpolating chapters 14–19 to commemorate King Alfred. Craster was interested in the Northumbrian Church of the Viking Age and needed the theory of an ‘early core’ to legitimize his reliance on HSC, but the hypothesis was not itself necessary to explain any problem offered by the text and it required ‘interpolations’ to be plausible, because HSC’s Alfred episode was clearly influenced by the late-tenth-century Vita prima Sancti Neoti, and because HSC’s author betrays his knowledge of the battle of Assandun (1016). Craster’s interpretation dominated views of HSC until the appearance of Ted Johnson South’s edition of the text in 2002. HSC’s modern editor has offered a simpler and more convincing case for a unitary text of eleventh century date, reinforcing the other scholarship that had already demonstrated stylistic and narrative unity to HSC. Footnote ²⁴

In the 940s there would have been people alive who could remember the ‘Flight of Eardwulf’, and a text written at that point would command some respect. However, even Craster himself disavowed such a simple approach to HSC. Although he did argue for an ‘early core’, he explicitly stated that ‘[i]t does not follow that the entire Cambridge text goes back to so early a date’.Footnote ²⁵ Craster’s vision of HSC would not, independently, give automatic credence to the ‘Flight of Eardwulf’. Since the HSC tradition encompasses a notice about gifts made by King Cnut (r. 1016–35), the earliest date for the completion of HSC (as we have it) would be that reign; with or without the ‘Flight of Eardwulf’, HSC as a unitary compilation cannot be given a meaningful date that precedes the reign. Although there would remain the theoretical possibility that HSC came from the first third of the eleventh century, there would be no way to establish a secure date even this early. All we can say for sure is that it was composed by the first decade of the twelfth century, prior to Symeon of Durham’s LDE and prior to Symeon’s inclusion of HSC in the Oxford manuscript.Footnote ²⁶ The ‘early core’ debate is, in fact, not central to how we should evaluate HSC. HSC did have access to early information (see below), but we do not know whether the ‘Flight of Eardwulf’ came from the type of source that would likely preserve early evidence reliably. In any case, no ‘early core’ case that relied on interpolations could ever coherently discount the likelihood of small-scale interference, like the type that could change the name of a church from, say, Cunecacestre to Ubbanforde. The most important observation to make is that our ability to date HSC is simply not good enough to draw the kind of general judgements we would need to establish a tenth-century date for any uncorroborated information with the text, let alone to infer confidently and specifically that the ‘Flight of Eardwulf’ represents an uncorrupted narrative pre-dating the ‘Resting-Places of Saints’.

To understand the value of HSC, we have to evaluate its chronological framework critically. Many Anglo-Norman era writers take their basic chronology from a known annalistic tradition, the Anglo-Saxon Chronicle; they are fundamentally reliable when they depend on this. It is reasonably clear, by contrast, that the architect of HSC did not always have adequate chronological guidance. Although the creator of HSC benefitted from, perhaps indirectly, Bede’s Historia ecclesiastica, early vitae of Cuthbert and the Anglo-Saxon Chronicle, otherwise he seems to have had little help figuring out Northumbrian chronology. By the time he brought HSC together he had not reconstructed the chronology well enough to realize how much time had passed between Cuthbert’s life and the ninth century. Among the more serious signs of the compiler’s limitations, seventh-century Cuthbert himself is presented as the direct predecessor of Bishop Ecgred, who, we know from elsewhere, held office in the ninth century.Footnote ²⁷ The compiler believed that Cuthbert (†687) had been a contemporary of King Ceolwulf (r. 729–37) and that he had been succeeded directly by the ninth-century bishop (chapters 7 and 8, with first lines of chapter 9). Similarly, he confused the eleventh-century battle of Assandune (Assandun) and ninth-century Ethandun (Edington), a point that may undermine the tradition’s credibility as a witness even to eleventh-century history.Footnote ²⁸ The confusion could be read to suggest that even the second battle had already fallen outside, or at least into the margins of, living memory; not very likely for the era of Cnut, but it could point to an incoming Continental or Southumbrian scholar relying on a mixture of oral and (possibly second-hand) written sources. English authors are, however, possible. It is worth remembering that native Northumbrian members of the Cuthbertine community experienced a period of marginality within their see for a significance chunk of the eleventh century. Since the early 1040s Southumbrian bishops (from Peterborough) had presided over Durham, an imposition on the community by the kings of England and the northern ealdormen based at York. The Norman Conquest saw the demise of this arrangement, along with a short period of resurgence for the native community; so HSC could also be explained as an attempt by the resurgent group to reassert their own ‘historical memory’, along with its concomitant privileges, even if, perhaps, it was a Norman audience they had in mind.Footnote ²⁹

There are other considerations that suggest HSC was drawn up during the post-Conquest period. It should go without saying that the labour, ink and animal skin used to produce HSC were not expended in a charitable attempt to help modern historians understand Viking-Age politics. HSC is a series of territorial and jurisdictional claims promoted through assertions about historical acquisitions and losses; it is best understood, above all, as asset advocacy. The creator’s achievement is to arrange Cuthbert’s procurements and injuries sequentially in accordance with the best chronology he was able to construct; further verisimilitude is provided by a patchy background taken from what little he was able to discover about Northern English history. In essence, it is a collection of pseudo-charters compiled for the purposes of proprietorial security and aggrandisement, meant to have an effect beyond the community at Durham. This type of pragmatic, contest-driven ‘historical’ writing is widespread in the eleventh century, and HSC shows many of the same Norman literary–legalistic tendencies illustrated by the pancartae especially common late in the reign of William the Conqueror.Footnote ³⁰

The specific territorial assets chosen for historical representation by HSC could lend some weight to the above interpretation. HSC is not simply a list of properties owned by the Cuthbertine corporation that the compiler has listed for historical interest; HSC focuses tendentiously on properties that were almost certainly not in their possession in the Anglo-Norman era. Although HSC’s dossier of asset rights included properties held in the eleventh century (like Crayke and Darlington), on the whole, its compiler seems to have been much more interested in targeting properties like those in the coveted Wapentake of Sadberge, or among the group of churches retained by the Northumbrian earls that, as their successor, King Henry I used to endow his new foundation at Carlisle.Footnote ³¹ Then there is Billingham, a ‘lost’ asset that according to LDE was successfully ‘regained’ by Durham after a grant by William the Conqueror. HSC’s author has included two apparently contradictory stories about how Cuthbert had lost this property. According to one account, it had been stolen by the notorious Northumbrian king Ælla; according to another, the Norse king Rǫgnvaldr, grandson of Ívarr, had been to blame. There are ways to explain this if you believe HSC is faithfully reproducing facts from the tenth century; but, realistically, this looks to be something like what lawyers call the ‘kitchen sink approach’, and that the author or the establishment at Durham had come across two reasonable arguments for repossessing the property, but was unable to decide which to use – perhaps with good reason, as King William may not have regarded one or the other, Ælla of Northumbria or Rǫgnvaldr of the Anglo-Danes, as his legitimate predecessor! William the Conqueror’s historic grant of Billingham might, then, tempt us to date HSC’s compilation to his reign; but that would be dangerous too, since the region beyond the Tees became politically unstable in the 1080s, creating opportunity to regain Billingham for anyone deprived by the Conqueror’s gift.Footnote ³²

USING HISTORIA DE SANCTO CUTHBERTO

HSC seems to be aimed at those able to grant or remove property rights or influence the related process, for example, the king, his court, the populace and so on. It is easy to account for this motivation in the colonial land grab that reshaped English society in the late eleventh and early twelfth centuries. Foreign conquerors were confiscating and redistributing land, and native property rights were frequently being challenged and lost to nobles and institutions of Continental origin (or even sometimes to their more adaptable native rivals). The process created a surfeit of charter histories and other historical writing by great churches trying to demonstrate the authority of their possessions and privileges. Other English episcopal churches produced similar accounts, including Hemming’s Cartulary and Liber Eliensis, comparable but more substantial texts from Worcester and Ely.Footnote ³³ As both Hemming’s Cartulary and Liber Eliensis illustrate, researchers in the Anglo-Norman era did seek out and utilize available evidence from written and oral sources. In Durham’s case, sources seem to have been in shorter supply, but we cannot doubt their existence. Separable components of HSC may, therefore, have potential as sources for the preceding two or three centuries. Indeed, attempting to dismiss the content of HSC entirely will create far more problems than it solves, particularly in regard to detail placed in the reign of Edward the Elder (r. 899–924). How, for instance, was the compiler of HSC able to invent an accurate synchronism of the Battle of Corbridge with several rulers and nobles of the ‘Edwardian’ era? There is no obvious known source for this information, but it is accurate (here we are able to verify the details from reliable texts).Footnote ³⁴ Again, what about chapter 33 of HSC, which is attested independently in a separate collection containing what appears to be late-tenth-century material?Footnote ³⁵ The charters purporting to be issued by King Æthelred II and Styr son of Ulf seem to be based on authentic exemplars, while the similarity of a grant attributed to Earl Northman in chapter 31 with notitiae found in Durham’s Liber Vitae could suggest that similar notitiae were used for some of the alleged later tenth- and early-eleventh-century grants.Footnote ³⁶

A charter or land grant on the margins of a gospel book could have provided some easily distilled and reproducible information with reliable synchronisms; the ‘Flight of Eardwulf’ on the other hand, an episodic narrative, does not jump out as this type of text. Like the Alfred episode, it may originate in some type of oral or hagiographic tradition and community myth, potentially one that took some time to emerge.Footnote ³⁷ That does not mean necessarily that the architect of HSC recycled a source that, in its original form, post-dated the tenth century; but it is difficult to employ the accompanying detail as a window on the Viking Age when we rely only on HSC and only on manuscripts from the twelfth century or later. It is not possible on current information to distinguish information in any hypothetical earlier text from what the ‘author of HSC’ could have added. Assuming for the sake of argument that the ‘Flight of Eardwulf’ was extracted from some earlier source, there would still be no particular reason to believe that the episode’s position within HSC’s chronological framework had any authority. How could we know that the creator of HSC was correct to synchronize the ‘Flight of Eardwulf’ with the reign of Alfred the Great? Again, even if we were to believe that ‘Ealdhun’s Translation’ had substance, how can we verify the date provided by Symeon? What allows us to believe that the compiler of HSC or even LDE was able to assign these episodes accurate synchronisms, dates or even eras?

SYMEON OF DURHAM AND VIKING-AGE CHRONOLOGY

It is worthwhile, here, to stress the limitations of the historical chronology available, more generally, at Anglo-Norman Durham, and to consider what we might be able to work out about its evolution. Historians writing in Anglo-Norman Durham did not begin with a systematic chronology integrating the history of their church with the wider world. That was something they had to construct. LDE is the closest they got to a final product. The author of LDE, responsible for the completion of this masterpiece and its accompanying chronological framework, may have relied on work carried out by a predecessor or with the help of associates. An earlier version of the ‘Symeonic’ chronology is attested in ALD, annals entered into margins of a manuscript of paschal tables (Glasgow, University Library, Hunterian 85) by the same hand identified today as Symeon’s.Footnote ³⁸ Those annals attempted to integrate the history of Durham with that of England and Christendom more widely, providing key dates of numerous bishops within a broader set relating to English kings, popes, emperors and so forth. Like LDE, ALD as a document has limited independent value for the Viking Age. While the author was able to draw on familiar Continental and English sources for the chronology of the aforementioned notables, there is no known single source that verifies his synchronization of events from Viking-Age Northumbria. Indeed, where Viking-Age material in the ALD can be tested or compared with other sources, even with sources likely to have been available to him, there are serious problems: Rollo’s takeover of Normandy is set to an implausible year 807, Æthelstan’s punitive expedition to Scotland set to 924 (recte 934), and Siward’s battle against Macbeth of Scotland set to 1046 (recte 1054).Footnote ³⁹ The attempt to establish a chronology for Durham history was clearly a work in progress. Reliable assistance was occasionally available, but the limitations of available source material forced them to make informed guesses at times. We have to accept that there is at least a very strong possibility that this applied to much of the Viking-Age Northumbrian material, which would include ‘Ealdhun’s Translation’; by extension, this would also include HSC’s treatment of the ‘Flight of Eardwulf’.

PRE-SYMEONIC CHRONOLOGY?

The author of LDE and his predecessors ultimately had to account for the fact that the diocese of Lindisfarne, the ancient see very famously documented in Bede and elsewhere, had by the time of the Conquest come to be re-seated far to the south, in Durham. The problem would have become an unavoidable headache when attempting to draw together a systematic chronology for the see. If we can find Anglo-Norman historical writing that pre-dates, or otherwise ignores, the solution to this problem offered by Symeon of Durham, we may have a better idea about how to contextualize and evaluate Symeon’s new chronological scheme. The Symeonic material is not the only product of early Anglo-Norman Durham, and it is worth recognizing that other sources sometimes provide detail at odds with the Symeonic system. One of these appears to be Cronica monasterii Dunelmensis [hereafter CMD], an early chronicle identified by Edmund Craster. CMD’s list of property acquisitions is organized according to a schema of English rulers very similar to HSC’s, but CMD’s is much more streamlined and without the same extreme grievance-driven political narrative. Craster argued that CMD had been completed in the time of William the Conqueror (†1087), during the episcopate of William de St Calais (1081–96); CMD records a grant that cannot predate 1085, but Craster also thought that this may have been added to a text written between 1072 and 1083 – meaning that its ‘core’ could be as early as the episcopate of Walcher (†1080). Written or transcribed into the margins of a book on the altar of Durham Cathedral, CMD came to be preserved in an appendix to a chronicle overseen by Prior John Wessington (†1451) and in a notarial instrument of 1433.Footnote ⁴⁰ In this case the chronological distance between composition and first manuscript witness is relatively large and must lighten the weight of evidence drawn from it; but if Craster is correct about CMD, we would have another window on history-making at early Anglo-Norman Durham that could provide an insight into the decision-making of Symeon and his associates.

One interesting point is that CMD does not seem to be aware of the critical ‘Flight of Eardwulf’ and, thus, does not mention Chester-le-Street or any move by Cuthbert to that location. If the historian were to follow Craster and take CMD as a witness to early Durham historical writing, CMD suggests that an otherwise well-informed researcher in Anglo-Norman Durham had no knowledge at all of Bishop Eardwulf or his deeds. Could this just be an editorial omission? After all, the inclusion of the Tyne–Tees grant in the ‘Donation of Guthred’ may suggest that the same contributor ‘knew’ about the Chester-le-Street move, but that it was omitted or removed for concision. On the other hand, the ‘Flight of Eardwulf’ would be an odd casualty of later textual interference given the significance of the story to Durham Cathedral after 1100. Why would Bishop Eardwulf specifically be ‘omitted’ from CMD’s version of the ‘Donation of Guthred’? The omission would make sense if the ‘Flight of Eardwulf’ was not part of the author’s framework for understanding the Cuthbertine house in the Viking Age, and that could mean that the move to Chester-le-Street did not gain a major role in Durham historical theory until later in the Anglo-Norman era. That is to say, it might suggest that there had been a pre-‘Flight of Eardwulf’ stage in Durham historical theory that involved ignorance of the Chester-le-Street move. The suggestion is supported elsewhere in CMD. When King Edmund (†946) is given his turn in the sequence of kings, CMD tells us that soon after his accession (939), the royal donor conducted an expedition to suppress Scottish incursions. When Edmund comes to pay his respects to the shrine of Cuthbert and to make the donations [that formed the real interest of CMD], he does so in Durham (Dunelmum), at the ‘church of Mary the mother of God and the holy confessor Cuthbert’ (‘ecclesiam sancte Dei genitricis Marie et sancti confessoris Cuthberti’).Footnote ⁴¹

Later Symeonic chronology is very clear that Durham did not become the seat of Cuthbert until decades after King Edmund; according to the familiar Symeonic schema, the seat should have been Chester-le-Street during King Edmund’s reign. Obviously, the alleged mid-tenth-century shrine at Durham has no direct implications for where the shrine actually was in the mid-tenth century – it would be natural for an early Anglo-Norman to locate the shrine at Durham if he knew nothing else otherwise. However, the architect of CMD took great care with his chronological structure, which would mean that the ‘error’ and the ‘omission’ of Eardwulf could (and if CMD is what Craster proposed, probably should) be taken as positive evidence to suggest that, even as late as the episcopate of William de St Calais, at least some of historical investigators, despite completing considerable research on the history of Viking-Age Durham, had not yet factored a Chester-le-Street theory into their chronology of the shrine’s location and that the ‘Flight of Eardwulf’ was a subsequent, and thus a relatively late, ‘discovery’.Footnote ⁴²

CMD is not the only early reconstruction of Cuthbertine history that fails to mention a Chester-le-Street stage. Descriptio status ecclesie Lindisfernensis et Dunelmi, another tract from the early Anglo-Norman era, mentions how Eardwulf and ‘several of his successors wandered hither and thither’ with the body, but makes no suggestion that Cuthbert’s relics went anywhere but Durham. The tract may date to 1083 and the episcopate of William de St Calais, when it ends; but a later date is plausible (indeed, Rollason suggested it was a later summary of LDE), and it would be possible in any case that the omission was one of editorial concision rather than ‘ignorance’.Footnote ⁴³ Yet another tract, known a little misleadingly as De obsessione Dunelmi, ‘On the Siege of Durham’, contains some potentially better corroborating evidence. This collection of historical material appears to be designed to support a Cuthbertine attempt to ‘repossess’ the estates of Barmpton and Skirningham. It cannot have been finished before 1073, but there is a possibility that it was completed not long afterwards, at least in its early form.Footnote ⁴⁴ The particular siege of Durham it purports to record is synchronized with King Æthelred II of England, Earl Uhtred, and Máel Coluim mac Cinaeda, king of Scotland; although the synchronism itself is historically possible, the absolute date that De obsessione Dunelmi provides, 969, is incompatible with the potential date range of the synchronism, 1005 × 1016.Footnote ⁴⁵ Although it appears to be a chronological miscalculation, the misjudgement offers insight about the understanding behind it. A scribal error that affected the century, decade and year must remain a theoretical possibility, but only an extremely unlikely one;Footnote ⁴⁶ but on the face of things, the author of De obsessione Dunelmi believed that there had been a bishopric at Durham in 969; this in turn would suggest that, like the compiler of CMD, he did not have Symeon’s chronology nor did he factor in the Chester-le-Street stage of Cuthbertine history that came to be so integral to the picture drawn up in the time of Symeon.Footnote ⁴⁷ Plausible alternative explanations can be given, no doubt, for these anomalies on an individual basis; but, to the extent it supports anything the cumulative picture seems to be more in line with a later rather than an early beginning for the ‘Flight of Eardwulf’ in historical theory at Anglo-Norman Durham.

DURHAM MATERIAL IN ANGLO-LATIN ANNALS

The ‘Flight of Eardwulf’ and ‘Ealdhun’s Translation’, along with certain other pieces about the Viking-Age Cuthbertine community, appear in some of the Anglo-Latin annals compiled in the twelfth century. These annals could be regarded as confirmatory if, for instance, they were early or if they revealed independent use of a source also used by HSC or Symeon (assuming they are distinct). The chronicle work in question consists of the northern Anglo-Latin annals contained in Historia regum ‘Part 2’ and its derivatives, which would include Historia regum ‘Part 1’ (for this information), as well as the annals of Roger of Howden’s Chronica, ‘Roger of Wendover’, and related sources. The common tradition carries an entry on the Eardwulf episode, placed s.a. 875, similar enough to HSC as to leave no doubt that there is, at least, a common source.Footnote ⁴⁸ The episode is built into what had originally been the equivalent year’s entry in the early Anglo-Saxon Chronicle tradition (as translated into Latin); the relevant entry is sandwiched between the notice of Hálfdan and a list of Norse leaders said to have wintered at Cambridge.Footnote ⁴⁹ Thus, it appears to be a genuine ‘interpolation’ originating at a later point. Another ‘interpolated’ note added s.a. 883, describes the end of the wandering and the reseating of the bishopric at Chester-le-Street.Footnote ⁵⁰ Most of these later versions of the account include Bishop Eardwulf and, like HSC, have a synchronization with King Alfred of Wessex. Most of them say that the body was moved to Chester-le-Street after many years of wandering, in most cases the figure being seven years.Footnote ⁵¹

The hope that these entries reveal a long, secure chain of transmission appears to be dashed by the earliest certain witness to this tradition of Anglo-Latin annals: Chronicon ex chronicis, currently attributed to John of Worcester and now known by his name.Footnote ⁵² In this compilation, the basis for Historia regum ‘Part 2’ and other annal collections named above, we find instead that a scribe writing sometime between 1128 and 1140 (or soon after) added a marginal note, s.a. 995, summarising the account about Bishop Eardwulf and the flight to Chester-le-Street, remarking that the body had remained there until moved to Durham in the time of King Æthelred II.Footnote ⁵³ This is one of a number of Durham-related notitiae added to Chronicon ex chronicis by later hands. Durham material in these annals originates in LDE, which makes it almost certain that there were no such Durham entries in that particular annal tradition prior to Symeon of Durham’s historical work; by extension, we can probably rule out the inclusion of these episodes in any pre–twelfth-century annals ancestral to surviving ones.Footnote ⁵⁴ At the very least, the annals as we have them do not offer independent authority for either the ‘Flight of Eardwulf” or ‘Ealdhun’s Translation’ as early sources. In essence, despite amplification by the annal tradition and modern historiography, the tradition about the move to Chester-le-Street (rather than Norham) in the ninth century seems to represent a single strand of historical evidence. A single error, misunderstanding, mix-up, rationalization, lie or act of narrative recycling, made perhaps c. 1100, could be responsible. As things stand, it is only this that needs to be weighed against the ‘Resting Places of Saints’ and William of Malmesbury.

‘FLIGHT OF EARDWULF’ AND THE PERSONAE OF DURHAM

Given the imperfect information available to the modern scholar, we are not in a position to treat the ‘Flight of Eardwulf’ as a reliable account of any event that happened in the Viking Age; even if the episode documents a real event, we have no particular reason to trust the chronological context built for it in the eleventh and twelfth centuries. None of that means, however, that the tale was invented by the compiler of HSC or by Symeon or his associates. We can allow a lot of room for rewriting, ‘re-spinning’ and creative storytelling, but the modern historian will face fewer issues by accepting that, most likely, the ‘Flight of Eardwulf’ is modelled on some Northumbrian English tradition, even if only an oral account. The same logic may apply to ‘Ealdhun’s Translation’, though the latter is so similar to the ‘Flight of Eardwulf’ that outright fabrication probably should be considered as a serious possibility. In any case, the limited value these accounts have for pre-Norman history should not be allowed to obscure one of the few things we seem to know: what HSC and LDE do tell us about the ‘Flight of Eardwulf’ is that, around 1100, it was a key story for the principal native stakeholders in the Cuthbertine corporation: the men claiming an inherited right to attend the body of Cuthbert.

These men, whom we can call personae (following conveniently ambiguous contemporary usage),Footnote ⁵⁵ claimed descent from ancestors who had, according to legend, personally guided the body of Cuthbert on its seven-year exodus from Lindisfarne to Chester-le-Street. LDE mentions four: Hunred, Sitheard, Eadmund and Franco.Footnote ⁵⁶ LDE further provides genealogies descending from two of them, Hunred and Franco: Hemming, priest of Sedgefield, and his brother Wulfkill, priest of Brancepeth, are said to descend from Hunred Cretel through their mother; a certain Ælfred, son of Alchmund the priest, is said to descend through his grandmother from Franco.Footnote ⁵⁷ One of Hunred’s descendants, Collan son of Eadred, was regarded as the first prepositus of Hexham in a twelfth-century tract on Hexham, taking up the office during the episcopate of Eadmund.Footnote ⁵⁸ As Symeon put it, ‘many of their descendants … take pride that their ancestors are said to have served St Cuthbert so faithfully’.Footnote ⁵⁹ The seven-year exile and supernaturally-guided foundational ancestors present the ‘Flight of Eardwulf’ as a ‘charter myth’.Footnote ⁶⁰ The senior members of the Anglo-Saxon Cuthbertine community who encountered the Normans in the eleventh century seem to have used something very similar to the ‘Flight of Eardwulf’ to account for their own status and privileges.

The ‘Flight of Eardwulf’ (and indeed ‘Ealdhun’s Translation’) bears noteworthy similarity to another creative origin myth of the era. The native establishment at St Andrews, the bishopric immediately to the north on the east coast of Britain, produced a tale with some striking similarities. The ‘B-version’ of the ‘St Andrews Foundation Legend’ recounted the flight of a certain Bishop Ríagal (Regulus) from Patras with pieces of Saint Andrew’s skeleton. In both aims and methods, this foundation legend is very similar to the ‘Flight of Eardwulf’: Hálfdan corresponds to the Emperor Constantius in causing the ‘flight’ and demise of the old order, Guthred is equivalent to the Pictish ruler Hungus for facilitating the emergence of the new order; the ecclesiastical father figures, Ríagal and his followers, build seven churches after landing in Fife. A larger number of companions of Ríagal are also named; but the seven churches, like the seven companions of Eardwulf, seemingly narrativize the corporate structure of a Viking-Age monastery or monastic familia. Indeed, according to another early St Andrews text, the so-called ‘Augustinian account’, there were seven personae at St Andrews c. 1140.Footnote ⁶¹ The story had probably been one of many similar collective origin legends propounded by Britain’s native ecclesiastical establishment around 1100.

Recognizing the ‘Flight of Eardwulf’ as origin or ‘charter myth’ circulating in Symeon’s time eliminates the need to debate the precise truth or falsity of the episode, at least in its extant textual form. For the hopeful historian, the detail presented by a such a story may ‘hark back’ to a precise set of real events, but as we saw above it is not possible to rely on the chronological framework offered by HSC or by Symeon even if this hope could be well founded. The mythic episode is offering us more about the politics of the era in which it emerged than the era with which it was synchronized by HSC and Symeon. Unfortunately, as we have seen, we cannot be certain about the context of its emergence because of the uncertainty about the date of all the components of HSC. In terms of the available evidence, it is possible that this myth emerged as late as the post-Conquest era as an attempt by the personae and their associates to gain credibility with their new Norman overlords or to counter disdain directed at them by incoming Benedictines; or perhaps they managed to use it to convince colonial incomers like Symeon that their origin legend should be regarded as a core part of the Cuthbertine shrine’s story prior to the institution of Southumbrian Benedictines and French bishops as the religious house’s dominant stakeholders. The episode’s lack of solid accompanying chronology would have given significant interpretational freedom to historical researchers like Symeon, but it would also have created pressure to reconcile solid chronology derived from written sources with the ‘truth’ of the ‘charter myth’ and related tradition. That would have included not only the Eardwulf episode itself, but satellite genealogical traditions relating to the ancestors of the personae.

If the historian looks closely at LDE, this pressure seems to have made its mark on Symeon’s project. There appears to be, for instance, a doubled-up parent-child relationship within a pedigree provided for one persona: its Hunred–Eadwulf–Eadred–Collan–Eadred–Collan lineage is the kind of thing that would have been more plausible in Symeon’s world than that of the modern specialist of Anglo-Saxon naming practices, where this pattern is rare. Symeon’s procrustean chronological bed could also explain the 210-year stretch of life given to a certain Riggulf, grandson of companion Franco, who had been part of the move to Durham from Chester-le-Street; again, this kind of lifespan might have been more plausible to the early-twelfth-century monk familiar with the ages assigned to early biblical figures than it would be to a modern scientist familiar with human ageing.Footnote ⁶² The important point here is that, when corrected, these pedigrees are incompatible with the narrative chronology that Symeon presents. Without the superhuman lifespans added by Symeon, these genealogies would take the personae ancestors back not much earlier than 1000, certainly not the ninth century.

Reginald of Durham indicates that another of the bearers of the body, (what in the ninth century would be) the anachronistically-Scandinavian name Eilaf, had been caught stealing cheese and transformed into a fox. Though Cuthbert returned him to human form, Eilaf’s descendants retained the name Tod, translated uulpecula (‘little fox’).Footnote ⁶³ Eilaf’s kin, Reginald claims, became holders of Bedlington (one of Durham’s exclaves in Northumberland) by hereditary right,Footnote ⁶⁴ and indeed one Eilaf is recorded as ‘of Bedlington’ in a purported charter of 1085.Footnote ⁶⁵ The rationalizing assumption would be that the latter Eilaf and the ancestor were distinct people, but by Reginald’s time in the later twelfth century the eleventh-century Eilaf could have come to be remote enough to telescoped with the ninth century as part of distant ‘deep past’.Footnote ⁶⁶ This Eilaf may even have been the same Eilaf who was priest of Hexham at the time LDE was composed. Other sources indicate that this office, although nominally conferred by the bishops of Durham, had been passed down to him from his father Ælfred, son of Westou, who himself appears to have acquired the right through marriage to a female descendant of Hunred. Eilaf Tod could, then, be an attempt to reinforce the position of Ælfred’s descendants by beefing up the historic credentials of their agnatic line.Footnote ⁶⁷

One cannot rule out, entirely, the possibility that the events of the ‘Flight of Eardwulf’ originated under the influence of an early textual source, but there is no confirmation of this from sources that are demonstrably early. There may have been some earlier basis for the account, and some earlier source may yet turn up; but, as things stand, we are not able to verify the existence of such a source, and we have to acknowledge that it remains based on hope and accompanying speculation. The episode seems to make sense as an origin or ‘charter myth’ of the personae of Durham, which would allow us to account for significance without necessarily assigning the account much usefulness as a guide to Viking-Age history. Indeed, as we saw above, there appear to some signs that Symeon’s attempts to reconcile it with his own chronological system caused him problems.

Whether or not that explanation is accepted, the move to Chester-le-Street in the later ninth century is not a historical event that is entitled to the credibility often given it by historians working on the era. We should seek other sources that may shed light on the Viking-Age episcopate in the far north of England, and interpret them with an open mind, without any particular need to reconcile anything ‘surprising’ with the unreliable detail or chronological framework presented to us by HSC and Symeon. It is with this approach that the value of the ‘Resting-Places of Saints’ stands out. If we recognize that we are dealing in probabilities based on the value of the evidence we actually have, it makes no sense to use the late information in the Durham tradition to reject the information provided by the ‘Resting-Places of Saints’. Its information about Cuthbert’s body could very well be out of date, but there is no better evidence for any alternative. Moreover, even if the ‘Resting-Places of Saints’ was out of date, that itself would not make the Durham material reliable. That is to say, the Durham account of the Cuthbertine community produced in the eleventh and twelfth centuries could not be used with confidence for the early Viking Age even if it was the only tradition that presented us with relevant dates and episodes. It is probably still open to question how reliable the ‘Resting-Places of Saints’ is for its information about Norham c. 1000, but the information provided by the ‘Resting-Places of Saints’, re-affirmed (if not corroborated) by William of Malmesbury, that Cuthbert lay at Norham c. 1000 prior to the move to Durham, is the closest thing we have to knowledge about the location of Cuthbert in the Viking Age.

MISCELLANEOUS EXTRACTS

As far as our problem in concerned, the value of HSC, and indeed William of Malmesbury and Symeon of Durham, is the potential to transmit earlier material. When there is reason to believe that a particular passage is earlier or originated separately from the surviving unitary account, historians must be prepared to review how the extract has been presented in regard to context and chronology. If a textual fragment or episode can be an independent source, then it can be evaluated independently—independent of a dating apparatus or narrative context provided by any compiler reproducing it. It is worth noting, then, that even HSC itself does reproduce a source supporting the ‘Resting-Places of Saints’ and William of Malmesbury. This is an extract that we can call the ‘Norham Account’: ‘Hoc tempore obiit sanctus Cuthbertus et successit Ezred episcopus, qui transportauit quondam ecclesiam olim factam a beato Aidano tempore Osuualdi regis de Lindisfarnensi insula ad Northam, ibique eam reedificauit et illuc corpus sancti Cuthberti et Ceolwulfi regis transtulit…’Footnote ⁶⁸ HSC as we have it is a unitary document, but some components of HSC were produced from pre-existing material. It happens to be the case that the compiler of HSC arranged the ‘Norham Account’ to precede the ‘Flight of Eardwulf’, but for us this would only matter if we were certain about the historical competence possessed by the architect of HSC. This person, as discussed above, thought that the ninth-century Bishop Ecgred had been the direct successor of the seventh-century saint (‘Hoc tempore obiit sanctus Cuthbertus et successit Ezred episcopus’).Footnote ⁶⁹

Unlike the ‘Flight of Eardwulf’, we do have evidence that the ‘Norham Account’ arose from a component of HSC that originated separately as a textual extract; and we also have some reason to believe that other historical writers of the Anglo-Norman era came to decisions about how to use the ‘Norham Account’ that differed from the decisions made by the architect of HSC. A variation of the ‘Norham Account’, arranged quite differently, is preserved in the later-twelfth-century Vita sancti Oswaldi regis, sometimes attributed to Reginald of Durham. The ‘Norham Account’ was included in this compilation because the compiler was concerned with the movement of Oswald’s head (which had accompanied Cuthbert). Vita Oswaldi’s version is more or less identical in words and detail to the version in HSC, except that the date given, 884, postdates the chronology suggested by HSC’s arrangement; and conflicts with the Symeonic framework.Footnote ⁷⁰ Vita Oswaldi as a whole need not concern us, but the vita rather than being a fluid hagiography is quite transparently a series of edited extracts that the compiler had been able to find in the work of earlier writers who discussed the life, death and relics of the saint.Footnote ⁷¹ This independent attestation of the story, seemingly presented in a separate chronological scheme, may provide another reason for believing that the detail of the ‘Norham Account’ could significantly predate HSC and the ‘Flight of Eardwulf’ (much like ‘Guthred’s Dream’, see below).

There are difficulties in assessing how the ‘Norham Account’ was incorporated into Vita Oswaldi. There is little reason to think that Vita Oswaldi is reproducing a Viking-Age source verbatim. The episode’s chronological setting is no less plausible than the alternative offered in HSC. Following Symeonic tradition, Bishop Ecgred is normally assumed to have died in the 840s, though this is not corroborated by contemporary sources. The later date offered by Vita Oswaldi, the year 884, still lies within the range of what is plausible,Footnote ⁷² and no obvious motive for fabricating it is immediately apparent. Nonetheless, it clearly contradicts the chronology and narrative finalized by Symeon and disseminated in the early twelfth century.Footnote ⁷³ The situation discourages the view that the otherwise rather clumsy author of Vita Oswaldi calculated the date himself, but there is no obvious reason to date the extract prior to the Norman Conquest. Theoretically, 884 may be an error, but it is difficult to dismiss or explain away with any type of reasoning that would not do even more damage to the credibility of the ‘Flight of Eardwulf’. Crucially, Vita Oswaldi suggests that whoever was responsible for the extract’s chronological framework believed that Cuthbert had been moved to Norham in the same year that Symeon had claimed (or would claim) Cuthbert had been moved to Chester-le-Street.Footnote ⁷⁴ Had this historian been influenced by William of Malmesbury, or by other established (or even subaltern) traditions about Norham? The modern scholar cannot know for certain, but our analysis hardly encourages confidence in the presentation of the Viking-Age past offered by HSC and Symeon. The author of Vita Oswaldi was able to reproduce a piece of text independently attested in HSC in a way that conflicts with its presentation in HSC: unlike the ‘Flight of Eardwulf’, the ‘Norham Account’ circulated in sources not created or influenced by Symeon. The simplest explanation is that the ‘Norham Account’ is based on pre-HSC and pre-Symeonic theory about Cuthbert’s Viking-Age movement, one that corresponds better to the contemporary evidence of the ‘Resting-Places of Saints’ than the familiar presentation of the ‘Flight of Eardwulf’. At the same time, although William of Malmesbury’s account is not a direct textual borrowing of the ‘Norham Account’ used by HSC, its picture is very similar to what is suggested by the ‘Norham Account’ in the context presented by Vita Oswaldi, which also points to access to a pre-existing source not subject (so the theory would go) to Symeonic revisionism.

According to HSC’s own particular arrangement of its extracts, Cuthbert was moved to Norham soon after his own death [sic] led to the succession of Ecgred (c. 9); later, during the time of King Alfred, Cuthbert’s body was relocated to Chester-le-Street. At face value, it might look as if HSC is suggesting that the corpse lay at Norham for a period in the early-to-mid-ninth century, until the last decades of the century when it was moved to Chester-le-Street. However, the detail is more complicated, and indeed even a face-value reading would not produce this coherent picture without ‘correction’ by a modern historian: HSC specifies that the body was moved from Lindisfarne (that is, not Norham) to Chester-le-Street.Footnote ⁷⁵ This difference may appear minor, but it is evidence that the architect of HSC has not properly integrated the ‘Norham Account’; the wording seems to be betraying the relative chronology of HSC. The modern historian has to invent a move back to Lindisfarne from Norham to rationalize HSC’s narrative. The two almost identical versions of the ‘Norham Account’, attested in HSC and elsewhere in Vita Oswaldi, happen to be arranged differently, but the modern historian is probably free to look at the ‘Norham Account’ and the standard version of the ‘Flight of Eardwulf’ as parallel, contradictory explanations for how Cuthbert left Lindisfarne, explanations which the compiler of the HSC may have attempted to reconcile merely through the relative positioning of each extract. Going further, if we believe that the compiler of HSC and Symeon were different people or had no contact with each other – even the former is far from certain, particularly as the handwriting in the earliest manuscript of HSC has been identified as Symeon’sFootnote ⁷⁶ – Symeon would have been faced with a similar choice between either the ‘Flight of Eardwulf’ or the ‘Norham Account’; in LDE he decided to favour the ‘Flight of Eardwulf’ and to discard the information about Cuthbert’s body presented by HSC’s ‘Norham Account’ (he only retained the part about the movement of Ceolwulf to Norham).Footnote ⁷⁷ Modern historians are not compelled to follow Symeon’s judgement here; and probably should not.

There are other items that should be mentioned not because they provide direct evidence against the Symeonic interpretation, but because they have consonance with the alternative picture suggested by the ‘Norham Account’, by William of Malmesbury and by the ‘Resting-Places of Saints’. One is yet another extract witnessed by HSC that appears, like the ‘Norham Account’, to have had an independent life. It is present in only one manuscript of HSC, and it is independently attested in a twelfth-century manuscript with tenth-century content, Paris, Bibliotheque nationale de France, lat. 5362 (fols. 53v–54r). South calls this extract ‘Guthred’s Dream’.Footnote ⁷⁸ In a battle between the Northumbrians and the Scots at a place called Mundingedene, ‘Guthred’s Dream’ relates that the Scottish host is swallowed by the earth. The precise site of Mundingedene has not been identified with certainty, but the surrounding action is set along the Tweed and, seemingly, close to Lindisfarne. Reginald of Durham in the later twelfth century described Munegedene as a hill on the Tweed lying in the vicinity of Norham – perhaps Ladykirk, but possibly the raised ground that later became the site of Norham Castle.Footnote ⁷⁹ The source does not mention the body of Cuthbert at Norham, but as a tale of potential tenth-century origin its depiction of a great threat to the Cuthbertine church and the lack of reference to anything further south than the Tweed basin would be in line with a tenth-century resting place at Norham (or indeed Lindisfarne).

Likewise, at the end of c. 21, HSC has a note of how a certain Tilred, abbot of Heversham (Westmorland), bought Castle Eden from Edward the Elder, giving half to Cuthbert ‘so that he might be a brother in his monastery’ (‘ut esset frater in eius monasterio’) and half to Norham ‘so that he might be abbot there’ (‘ut ibi esset abbas’). This could be another case where the HSC compiler may be preserving earlier material. It shows Tilred gaining membership of the Cuthbertine familia and an abbacy; it does not explicitly state that the ‘abbacy’ at Norham had particularly high status within the familia, but freed from the Symeonic straight-jacket that would be the natural inference to draw. If this information did come from an earlier source, the incidental notice of Norham in this type of context would strongly support the idea that it had been the centre of the Cuthbertine familia in the tenth century. If Norham was at the heart of the Cuthbertine see in Tilred’s time, it makes sense for an extract like this to have survived in Cuthbertine records. The appearance of Tilred’s name on the Cuthbertine episcopal lists, perhaps, strengthens the appeal of this reading, even though these only begin appearing c. 1100.Footnote ⁸⁰

As we saw above, material that was derived from HSC, probably via LDE, was re-used by being added to the northern Anglo-Latin annals derived from Chronicon ex chronicis, a tradition represented best by Historia regum ‘Part 2’ and Roger of Howden. The northern annals also utilize a number of additions from other source(s). Among these, there is a passage explicitly about the Viking-Age diocese of Lindisfarne, which for the sake of analysis we can call ‘Properties of the Diocese of Lindisfarne’. The extract has been inserted alongside the ‘Flight of Eardwulf’, and in this surviving form appears to be pushing a claim to Carlisle. Carlisle was a concern of Durham ‘historians’ in the late eleventh and early twelfth centuries, before King Henry I decided to make the site the centre of a new diocese. The geography and place-names used in the text, however, suggest the influence of an older exemplar, and indeed that the ‘Properties’ could be a recycling of a genuine list of churches subject to Lindisfarne in the Viking Age.Footnote ⁸¹ After proclaiming its ‘ancient’ possession of Carlisle and overlordship over all churches north of the Tyne, it lists a number of minsters subordinate to the bishopric of Lindisfarne in and around the Tweed basin. The list begins with Ubbanford / Norham, and proceeds around the moor moving from the lower Tweed into Teviotdale (Carham et Culterham, Jedburgh, Melrose), up into West Lothian (Tigbrechingham and Abercorn), through Midlothian and East Lothian (Edinburgh, Pefferham, Auldhame, Tyninghame) back to the lower Tweed (Coldingham, Birgham [only Howden], Tillmouth), returning to Ubbanford / Norham.Footnote ⁸² The ‘Properties of the Diocese of Lindisfarne’ reads as if the episcopal monastery of Lindisfarne claimed supervisory rights over the other ‘head ministers’ of the region between the Coquet and the Firth of Forth, but the presentation seems to be modelled on a circuit or itineration performed by a bishop based in Norham. In the early medieval Insular World, performance of a circuit was ‘a normal expression of lordship’.Footnote ⁸³ The extract could be either an accurate description of the Viking-Age diocese or else a tendentious claim made to advance Cuthbertine control over these churches at a later date, perhaps in the eleventh century when the political order of northern Northumbria was disrupted by Scottish, Danish and Norman invaders (see below). On the other hand, the description presents the diocese in a way that clearly pre-dates any merger with Hexham or incorporation into the greater diocese of Durham. Moreover, the centrality of Norham in the account appears as incidental, almost inconvenient information, and concurs with what the ‘Resting-Places of Saints’, William of Malmesbury and the ‘Norham Account’ suggest about the Viking-Age diocese of Lindisfarne and the location of its principal shrine.

EMERGENCE OF DURHAM: SECULAR EXPLANATIONS

The evidence considered above, the ‘Resting-Places of Saints’ and William of Malmesbury in particular, puts the shrine of Cuthbert at Norham in the early eleventh century, with the ‘Resting-Places of Saints’ suggesting that it was located there until at least 1013. For what it is worth, the move of the shrine to Norham is not a theory that necessitates the abandonment of monastic life in Lindisfarne, nor even that Lindisfarne ceased to be at the centre of the Cuthbertine diocese.Footnote ⁸⁴ A reliable Anglo-Latin annal tells us that a Hiberno-Norse army from York attacked Tyninghame as well as the community of Lindisfarne around 941, enslaving some of the population; sculpture on the island seems to reach a ‘second peak’ in the tenth century.Footnote ⁸⁵ The relocation of the shrine fourteen or so miles away at Ubbanford does not necessarily suggest any drastic reorientation. The site would have been slightly, but hardly much less accessible to opportunistic piratical predation; and Ubbanford was well located as a stop-off place on long-distance land routes. Lying just beyond the tidal limit of the Tweed, Norham is adjacent to the lowest fords over the river, and was probably better placed as an administrative centre for most of the surrounding Tweed basin than Lindisfarne, particularly if Scandinavian dominance of the open waters made Lindisfarne (far from being centrally positioned within the diocese) a somewhat riskier place to visit for native Northumbrians and for pilgrims coming from Southumbria or the Celtic-speaking regions.Footnote ⁸⁶

Despite the loss of much of Northumbria to Scandinavian settlement in the late ninth and early tenth centuries, some of the earlier kingdom of Northumbria appears to have survived as a rump, continuing to provide a set of rulers distinct from the Scandinavians who established themselves further south after the 870s. By the early tenth century, northern Northumbria was ruled by a certain Eadwulf from Bamburgh; his sons and descendants, the Eadwulfings, seem to dominate the far north of England until at least the time of Siward (died 1055). Although Southumbrian English sources deny them royal titles, two are described as ‘King of the Northern English’ by Irish annals in the early tenth century, and a late-tenth-century Scottish source refers to the capture by Cinaed son of Máel Coluim (died 995) of a ‘son of the King of the English’ during an invasion of Northumbria (‘predauit Saxoniam et traduxit filium regis Saxorum’), who cannot have been the son of the reigning [Southumbrian] English monarch Æthelred.Footnote ⁸⁷ Despite the establishment of a distinct ealdorman for the Northumbrian Anglo-Danish regions by the reign of Edgar, the more distant ‘Northern English’ principality was almost certainty left by the West Saxon rulers, at least for the most part, to its own devices. Charter attestations seem to suggest that both the rulers of Bamburgh and the Cuthbertine bishops were very remote from their power, and the connection of the region with the West Saxon dynasty seems to have been more like that of Strathclyde or Gwynedd than Yorkshire. Around 970, one Eadulf Dux and the Cuthbertine bishop, Ælfsige, both show up in Southumbrian documents.Footnote ⁸⁸ Almost a quarter of a century later in 994, another Eadwulfing, Waltheof (father of Uhtred) appears in Southumbria in the aftermath of the sacking of the Northern English capital of Bamburgh by a Scandinavian army.Footnote ⁸⁹ The next bishop thought to be associated with northern Northumbria (Ealdhun) does not make his solitary appearance until 1009 – a significant fact, given that the body of surviving charters is substantial at this stage and that the holders of most English bishoprics, including York, attest charters regularly.Footnote ⁹⁰

There can be no doubt that the diocese of St Cuthbert in the tenth century was very closely tied to the Eadwulfings of Bamburgh. It is even possible that the see was the single, ‘national’ bishopric of the Eadwulfing principality in the tenth century, in the same way the episcopus Scottorum / ardepscop Alban (early-twelfth-century titles for the bishop of St Andrews) was responsible for the Scottish kingdom. The negative imprint of Scandinavian place-names would suggest that the Eadwulfing polity was confined to the region a little north of the Tyne, with core territory in the Tweed basin but probably some intermittent power as far south as the Tyne as well as north over the Lammermuir into Lothian proper.Footnote ⁹¹ The territory described by the ‘Properties of the Diocese of Lindisfarne’ corresponds with this pattern somewhat. It is worth pointing out that in HSC there is another, less extensive description of the boundaries of the see where the Coquet appears to form the southern frontier of Lindisfarne’s jurisdiction.Footnote ⁹² Interestingly, that kind of frontier also seems to have been recognized in one of the historical traditions that Symeon of Durham included in relation to ‘Ealdhun’s Translation’. According to LDE, when the body of Cuthbert was moved to Durham, ‘people from the whole area between the river Coquet and the river Tees’ (‘a flumine Coqued’ usque Tesam uniuersa populorum’) came to help clear the vicinity of Durham of trees and construct a new cathedral.Footnote ⁹³

This geography would indicate that, unlike a move to Norham, a move to Durham was a radical change that would have significantly damaged the prestige and long-term position of the Eadwulfings (who were still very much around in the 1020s and 1030s). It may be no coincidence that the 1010s witnessed two disasters for them: the Danish invasions and the defeat by the Scots at the battle of Carham c. 1018, one of which led to the death of Uhtred.Footnote ⁹⁴ It was a high fall, because Uhtred had enjoyed a historically powerful position: in addition to ruling the northern Bamburgh polity, he married the daughter of King Æthelred and acquired authority further south, holding the viceregal ealdordom in [southern] Northumbria.Footnote ⁹⁵ The loss of this most prestigious, core shrine from the Eadwulfing polity could, then, be interpreted to mean a significant, sudden (if perhaps temporary) diminution of power in the aftermath of Carham and the Danish conquest of England. Cnut replaced Uhtred as ealdorman in York with a Scandinavian dux named Erik, and in the subsequent decades the Danish regime may have used its strength to reorder northern matters at the expense of the Eadwulfings. Cnut’s period as principal ruler of England and the Scandinavian world does seem to have stimulated changes to episcopal authority and diocesan structure, both in Scandinavia itself and in the ‘peripheral’ Scandinavian areas of the Insular world (for example, Dublin and Orkney, and, possibly, Glasgow).Footnote ⁹⁶ On the other hand, LDE claimed the move to Durham had been organized by Uhtred himself, a statement that is potentially baseless but which we are not free to dismiss entirely. Uhtred had exercised power throughout Northumbria as both ruler of Bamburgh and ealdorman in York and may have wished to tie the two regions back together in the interests of patrimonial aggrandizement, perhaps trying to ‘reunite’ Northumbria; and, presumably, he would have sought to do this before being deposed as ealdorman by Cnut in 1016. On the other hand, the role given to Uhtred by Symeon may also be conjecture, perhaps a side-effect of Symeon’s semi-speculative chronology or perhaps to deflate claims by northerners that the people south of the Coquet had stolen their relics unjustly.

EMERGENCE OF DURHAM: ECCLESIASTICAL EXPLANATIONS

There may also be possible explanations that are more ecclesiastically specific. Understanding the move from Lindisfarne to Durham is inevitably tied up with how we understand the Northumbrian episcopate in the Viking Age. We know from reliable early medieval sources that Northumbria had four bishoprics in the early ninth century: York, Hexham, Lindisfarne and Whithorn. Of these only only York survived by the Anglo-Norman era, alongside one other diocese with its seat at Durham. Far to the south of Norham and Lindisfarne, Durham looks more like a successor of Hexham than Lindisfarne. Indeed, despite its appropriation of the Lindisfarne’s relics and possessions, prior to the Viking Age the site of Durham was very likely within the boundaries of the diocese of Hexham. In the Anglo-Norman era, historians like William of Malmesbury trying to document the history of the English episcopate, found, as we today find, that early medieval episcopal lists cover Hexham and Whithorn only as far as the early ninth century. William and his modern successors read the pattern to suggest that the bishoprics of Hexham and Whithorn came to an end in the ninth century. Viking-Age episcopal lists covering Northumbria did indeed end in the early ninth century. But it is important to recognize that these cessations affected Lindisfarne as much as Hexham or Whithorn, and simply reflect the date the lists had been compiled. What does matter for the subsequent picture is that by 1100 episcopal lists had emerged that linked the see of Durham to the bishops of Lindisfarne. This is why William felt the need to account specifically for Hexham and Whithorn (but not the Cuthbertine see): in William’s case, he attributed the disappearance of Whithorn to incursions of the Scots and ‘Picts’, and Hexham to the Danes.Footnote ⁹⁷ William’s logic makes sense in Hexham’s case. After all, the ninth century had been the height of the ‘Viking Age’, when Scandinavian armies cleared the way for the ‘birth of England’ by sweeping aside much of pre-ninth-century political order; but the anachronistic incursions of ‘Scots’ and ‘Picts’ reveals William to be engaging in, or at least picking up, speculation that carries no authority.

The disappearance of the pre-Viking-Age episcopate in Northumbria has usually been regarded as one of the more definitively chronicled examples of reconfiguration in the episcopate of Viking-Age England.Footnote ⁹⁸ Again, this is another topic where Anglo-Norman-era historical narratives, and the survival patterns of Viking-Age evidence that the creators of these narratives encountered, have shaped modern understanding more than is merited. The demise of the classical Northumbrian episcopate in the Viking Age is not in doubt, but we have to be realistic about the kind of sources available and what kind of precision they offer for the chronology of this change. As far as contemporary evidence is concerned, we do not have to wait until the eleventh century for a window on the Northumbrian episcopate; that is provided, in fact, by charter attestations from King Æthelstan’s time as ruler in Northumbria.Footnote ⁹⁹ As a particular example, a genuine witness list from a charter in the Worcester archive, S 401, has Archbishop Hrothweard of York appear alongside four bishops explicitly stated to be his suffragans, with: Rodeward quoque archipræsul cum Eboracensis suffraganeis . Æscber’h’to . Wigredo . Earnulfo . Columbano . consignauit. Footnote ¹⁰⁰ While it is theoretically possible they were diocese-less assistant bishops to the archbishop, this is not particularly likely and indeed the third-ranked Wigred’s appearance in later Durham episcopal lists suggests that this is not the explanation. Very conveniently, there are enough empty bishoprics to account for Lindisfarne, Hexham and Whithorn, as well as one other bishopric.Footnote ¹⁰¹ The witness-lists of these charters do not, unfortunately, name bishoprics and do not rule out the movement of episcopal seats or saints’ shrines; but they do suggest that the number of bishops north of the Humber was, if anything, higher in the early 930s than it had been in the early ninth century. The reduction in Northumbria’s episcopate, the disappearance of component bishoprics, seems to have happened not in the ninth century, but at some stage after the 930s; it almost certainly had taken place by the time of the Norman Conquest, but it is difficult to arrive at more precision without relying on speculation.Footnote ¹⁰² Durham had probably become an episcopal see by the end of the episcopate of Eadmund (II): Æthelric of Peterborough, seemingly the first bishop of Durham styled as such in contemporary sources, is said to have taken the see in 1041 and to have relinquished it in 1056.Footnote ¹⁰³

It is worth noting, however, that CMD purports to reproduce a letter that could provide some help. The letter was sent by Bishop Eadmund to his colleague Ælfric of Winchester, archbishop of York (consecrated c. 1023). If genuine (Craster viewed it as an interpolation from the reign of Henry I, when control of the church of Hexham was a live issue), the letter would probably date to the reign of Cnut, and would indicate that Eadmund and Ælfric were disputing the legacy of the see of Hexham: Eadmund uses information drawn from Bede’s Historia ecclesiastica to show that York had no right to the control of Hexham’s diocese.Footnote ¹⁰⁴ Assuming for the sake of argument that the letter is not a forgery, it would appear to confirm that Hexham had ceased to be an episcopal centre prior to the establishment of a see at Durham, and that no clear successor to the diocese of Hexham continued to exist during Eadmund’s early episcopate.Footnote ¹⁰⁵ If so, the movement of the shrine south in the time of Bishop Eadmund could also be read as a firm assertion that the see originally based at Lindisfarne had adopted an expanded remit. The tenth-century Cuthbertine see had probably been acquiring properties in the Danish-settled areas that later became Yorkshire and County Durham; if Bishop Eadmund had moved the shrine to this area of Durham, it might have been to consolidate this position and head off attempts from York to annex the former diocese of Hexham (in the same way that the Great Northern War famously led to the foundation of a new Russian capital at St Petersburg).Footnote ¹⁰⁶ The possibility is at least worth considering, particularly for anyone who believes that the Cuthbertine corporation was autonomous and not merely a component of Eadwulfing political armoury. As a downside, that would make the significance of Chester-le-Street in the traditions reproduced in later writings more difficult to explain. Alternatively then, if Chester-le-Street was the successor of Hexham (see below), re-foundation at Durham may have been seen as a way to create a fresh but common centre for both sees.

Despite Symeon’s belief in ‘Ealdhun’s Translation’, he does reproduce one tradition that confirms Bishop Eadmund’s era as one of relic relocation. Symeon notes an existing historical anecdote in which Ælfred son of Westou, thesaurarius of Durham and holder of the church of Hexham, had had the bodies of major Northumbrian saints relocated to Durham from Coldingham, Melrose, Tyninghame, Jarrow, Tynemouth and Hexham. This relic relocation, Symeon learned, had happened during the episcopate of Bishop Eadmund.Footnote ¹⁰⁷ If Symeon’s tradition about Eadmund is reliable, it would indirectly support William of Malmesbury’s idea that Bishop Eadmund had presided over significant movements of Cuthbert’s relics. Whether motivated by security concerns or by opportunism, both sources suggest that Eadmund’s episcopate saw a concerted attempt to centralize the ecclesiastical affairs in northern Northumbria and create a single centre for the major relics of the old Bernician church, in a location further south and closer to York and ‘royal England’.

DIOCESE OF DURHAM IN THE ELEVENTH CENTURY

If we accept the reconstruction above, ecclesiastical lineages north of the Tyne must have suffered a significant diminution in prestige and in public power as they were shifted into a new periphery. Weaker localities are, of course, the inevitable side-effect of most processes of centralization; in this case, however, it is not particularly evident that the new church establishment at Durham retained control of the area previously covered by the Lindisfarne diocese based at Norham. In secular affairs, the ealdormen of York may have enjoyed loose overlordship of the land north of the Tyne for much of the eleventh century. Siward appears to have enjoyed particularly wide powers, premised partly on the death of the Bamburgh ruler ‘Earl’ Eadwulf and marriage to the latter’s niece, Ælffled.Footnote ¹⁰⁸ This power probably did not long endure after Siward’s death, and certainly did not survive the Norman Conquest; and there is no particular reason to see substantial unity restored to eastern Northumbria prior to the era of the last Norman earl, Earl Robert de Mowbray (deposed 1095). Among Robert’s predecessors, the earl–bishop Walcher (died 1080) had been unable to control much to the north of Durham itself. Indeed, the famous Southumbrian ‘monastic revivalists’ established at Melrose during Walcher’s era, men who, significantly, attended the deathbed of Walcher’s rival, Earl Gospatric, at Norham, lay outside Norman power, among the still free Northern English, until they were induced southwards by letters and pleas.Footnote ¹⁰⁹

Assuming for the sake of argument that Lindisfarne’s relics were relocated to Durham during Eadmund’s episcopate, we are left wondering why Eadmund’s alleged predecessor Ealdhun came to be regarded as the founder of the see of Durham. According to De obsessione Dunelmi, Bishop Ealdhun was regarded as the ‘original’ holder of a variety of properties in the County Durham area.Footnote ¹¹⁰ One of the other things we seem to know about Ealdhun is that he was part of the kin-group that came to rule over Hexham: the prepositus of Hexham confirmed by Bishop Eadmund was explicitly said to have been the nepos of Bishop Ealdhun.Footnote ¹¹¹ The southerly personae we know about look more likely to have been connected to Hexham than Lindisfarne, but it is also worth pointing out that St Cuthbert was remembered in connection to both sees, and could be presented as a source of unity for both dioceses. The personae of Cuthbert in the later eleventh century seem to be confined to territories south of the Coquet; if, as seems likely, these men held their offices as family honours, they are unlikely to have been the descendants of actual custodians of Cuthbert in Lindisfarne or the mainland Tweed basin; or, if they were, they lost whatever territorial base their ancestors had in the north by the time of the Normans.

There are other possibilities for this anomaly: the personae may have originated as ‘carpetbaggers’ who came from the north; similarly, they may have been southerly members of the Cuthbertine familia who survived the political turmoil of the early Viking Age. A reliable tradition in HSC does suggest, after all, that a minster as far south as Heversham supplied an incumbent for the abbacy of Norham in the tenth century.Footnote ¹¹² Another possibility is that Ealdhun’s kin had been hereditary holders of Hexham since it lost episcopal status but were also major stakeholders in some other church that succeeded Hexham. Could this Hexham successor have been Chester-le-Street? If so, that could mean that the ‘Flight of Eardwulf’ originated as a foundation myth for the personae of Chester-le-Street, subsequently applied to the greater honour after Ealdhun became bishop of St Cuthbert. In this scenario, Ealdhun may have been the bishop of Lindisfarne at Norham prior to Eadmund’s episcopate, before the relics themselves left Norham. This would leave open how, when and why Durham as a site became an episcopal centre rather than the location for Cuthbert’s relics. The temporary unity that Uhtred brought to Northumbria could have facilitated Ealdhun’s appointment to Lindisfarne/Norham, a move which would probably have brought Cuthbert more southern property, perhaps giving Eadmund even more of an incentive to move south, and cemented Cuthbert’s hold on these gains and other rights within the Danelaw.Footnote ¹¹³

The age of Uhtred and Cnut, and the time of Walcher, Gospatric, and Robert de Mowbray, were separated by several generations and more than a few forms of political order. In the intervening period, hereditary ecclesiastics associated with churches far to the south of the Coquet, and beyond, over the Tyne, at places like Hexham, Chester-le-Street, and Durham, had several generations to make history more ‘useable’ for themselves by coming to attribute greater chronological depth to what was, possibly, their much more recent association with northern England’s most prestigious set of relics. This could explain why Symeon of Durham and other ‘researchers’ at Anglo-Norman Durham might have chosen to overlook reliable information about the movement of relics from Norham to Durham in the first half of the eleventh century.